Skip to contentSkip to conversation

how to measure scheduler latency on powerpc? realfeel doesn't work due to /dev/rtc issues

DISCUSSION Peter Chubb 2 participants 2 messages May 12, 2003 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"William" == William Lee Irwin, <William> writes:

William> Not at all. Just stamp at wakeup and difference when it runs.

That then doesn't include interrupt latency. The nice thing about the amlat tests is that the test predicts when the next interrupt should occur, then measures the time between that prediction and the process running in userspace. If you just timestamp at wakeup, you miss all the time between interrupt generation and noticing that the process is to wake up.

--
Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au
You are lost in a maze of BitKeeper repositories, all slightly different.

#2 lore raw

"William" == William Lee Irwin, <William> writes:
William> Not at all. Just stamp at wakeup and difference when it runs.

Quoted 6 lines · Peter Chubb

That then doesn't include interrupt latency. The nice thing about the amlat tests is that the test predicts when the next interrupt should occur, then measures the time between that prediction and the process running in userspace. If you just timestamp at wakeup, you miss all the time between interrupt generation and noticing that the process is to wake up.

Of course. But that is not the scheduler's problem.

-- wli